RSCO Technical Discussion

[bhac0203]

I am currently running RSCO ads, but I’ve been unable to scale up the ad spend, so I wanted to start a thread to discuss this.
On June 11, the ROI was 1.03.
If anyone else is working on this, I’d love to exchange ideas. Thanks.

 

[bhac0203]

On June 10, the ROI was 1.22.
On June 9, the ROI was 1.12.
On June 8, the ROI was 1.18.
On June 7, the ROI was 1.5.

Starting on the 7th, the ROI began to drop. The reason for the decline was my attempt to scale up spend; I increased the budget, and as a result, the ROI fell to around 1.

 

[Rankings Daily]

yeah that drop makes sense honestly, you scaled too hard too fast. every time you bump budget by a big chunk the algo basically gets kicked back into learning and the early data is garbage, so ROI craters for a few days before it settles. problem is people see it crater and panic, but you can also be feeding it right when its still re-optimizing.

what worked better for me is going slow on the vertical side, like 15-20% bumps max and then leave it alone 2-3 days. anything more and you reset everything.

honestly tho the better move at your scale is to stop just raising budget on the one winner. duplicate the winning adset and run it in parallel, horizontal scaling. keeps your original stable and you test fresh audiences without nuking the performer. some networks the dupes behave weird but it usually beats forcing one set to spend 3x.

one more thing nobody mentioned... check your frequency before you blame the budget. you pushed the same creative since the 7th right? if freq is climbing your ctr is dying and cpa goes up regardless of how clean your scaling is. could be creative fatigue masking as a scaling problem. swap in 2-3 new angles and see if the 1.0 floor lifts before you mess with spend again.

 

[thebestcodeseller]

Hi

 

[bhac0203]

yeah that drop makes sense honestly, you scaled too hard too fast. every time you bump budget by a big chunk the algo basically gets kicked back into learning and the early data is garbage, so ROI craters for a few days before it settles. problem is people see it crater and panic, but you can also be feeding it right when its still re-optimizing.

what worked better for me is going slow on the vertical side, like 15-20% bumps max and then leave it alone 2-3 days. anything more and you reset everything.

honestly tho the better move at your scale is to stop just raising budget on the one winner. duplicate the winning adset and run it in parallel, horizontal scaling. keeps your original stable and you test fresh audiences without nuking the performer. some networks the dupes behave weird but it usually beats forcing one set to spend 3x.

one more thing nobody mentioned... check your frequency before you blame the budget. you pushed the same creative since the 7th right? if freq is climbing your ctr is dying and cpa goes up regardless of how clean your scaling is. could be creative fatigue masking as a scaling problem. swap in 2-3 new angles and see if the 1.0 floor lifts before you mess with spend again.

Thank you very much for your reply.
My current workflow involves adding new ad groups, creatives, and budgets. Although the new creatives feature different content from the old ones, adding these new ads impacts the account's overall budget—which is the most frustrating part.
My understanding is that adding new ads—especially those targeting different content areas—shouldn't affect the performance of existing ads; even if the new ads underperform, the old ones should remain stable.
However, I’ve noticed that when I add new ads, the performance of the old ones deteriorates as well.
I just checked the account's impression frequency; the overall repetition rate is around 1.3. This might be causing the audience reach to be too narrow, preventing the ads from breaking out to a wider audience.
By the way, for my RSOC campaigns, I’m targeting the "All Audiences" segment without applying any specific interest filters. Could this be the reason why my ads are only being shown to a small subset of the population?
Would it be better if I applied some audience filtering?

 

[bhac0203]

Hi

Hello there. May I ask if you are also working on RSOC operations? If so, we could exchange experiences. Since you only sent a "hi," I’m not sure what specific issue you have. Thanks.

 

[bhac0203]

I will continue to update this post. Friends who are running rsocs are welcome to follow and discuss it.

My testing over the past few days:

I tested with the same keywords using a new ad format. The results were the same: both the new and old ads performed poorly.

I tested with different keywords in the same market segment using a new ad format. The results were the same: both the new and old ads performed poorly.

I tested with different keywords in different market segments using a new ad format. The results were the same: both the new and old ads performed poorly.

I tested with different keywords in different countries in different market segments using a new ad format. The results were the same: both the new and old ads performed poorly.

My feeling is that I'm stuck at my current budget. Adding more ad budget, even for different countries, will negatively impact the existing ads.

If I maintain my current budget and don't add any new ads, my existing ad group's ROI stabilizes at around 1.3.

I have a question: Is it possible that a feed account, especially a new account, has a limited traffic flow in the early stages? Is it only possible to increase the traffic flow after the traffic quality passes verification and review?

 

[uniqueuser1235]

I am currently running RSCO ads, but I’ve been unable to scale up the ad spend, so I wanted to start a thread to discuss this.
On June 11, the ROI was 1.03.
If anyone else is working on this, I’d love to exchange ideas. Thanks.

Hi! I've been following your updates, and I think I understand what you're running into. I've spent a lot of time working with RSOC campaigns and have experienced similar budget limitations and scaling problems.

I don't know if my approach will solve everything, but I'm happy to share what's worked for me and exchange ideas. If you're open to it, what's the best way to contact you privately?

 

[bhac0203]

Hi! I've been following your updates, and I think I understand what you're running into. I've spent a lot of time working with RSOC campaigns and have experienced similar budget limitations and scaling problems.

I don't know if my approach will solve everything, but I'm happy to share what's worked for me and exchange ideas. If you're open to it, what's the best way to contact you privately?

Hello, I’d be happy to chat with you, but my account doesn't allow me to send you a private message. Could you send me one instead? Thanks a lot.

I’ve been running tests over the past few days—trying out various creatives and keywords—as I look for ways to break through the current spending limits.

 

[woolfreylyu145931]

Hello, I’d be happy to chat with you, but my account doesn't allow me to send you a private message. Could you send me one instead? Thanks a lot.

I’ve been running tests over the past few days—trying out various creatives and keywords—as I look for ways to break through the current spending limits.

Are you running feeds or direct AFS? As far as I know, feeds are not able to generate positive return at scale at all.

 

[bhac0203]

Are you running feeds or direct AFS? As far as I know, feeds are not able to generate positive return at scale at all.

I'm running feeds. My biggest problem right now is that I can't scale up the volume. I can run a positive ROI, but I can't scale up the volume.